© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 34-GEMMASCOPE-RES-131K
    7. 55688
    Prev
    Next
    INDEX
    Explanations

    new video, meeting, question, thoughts

    np_acts-logits-general · gemini-2.5-flash-lite

    This neuron fires on programming‐style tokens—especially underscores and numeric literals—i.e. bits of source‐code syntax.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_34/width_131k
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     cdr
    -0.81
     Hata
    -0.80
    valg
    -0.80
    Alla
    -0.77
    Wnd
    -0.77
    subnet
    -0.76
    iems
    -0.76
    宮城
    -0.75
    udt
    -0.75
     adjud
    -0.74
    POSITIVE LOGITS
     Nintendo
    1.04
     Wii
    0.85
     console
    0.85
    mont
    0.81
     confer
    0.79
     периода
    0.78
     erwarten
    0.78
     Mont
    0.77
     FIFO
    0.75
     display
    0.74
    Activations Density 0.015%

    No Known Activations